点击显示 收起
【摘要】 Sodium bicarbonate cotransporter 1 (NBC1) mediates 80% of bicarbonate reabsorption by the kidney, but the molecular determinants for activity, targeting, and cell membrane stability are poorly understood. We generated truncation mutants involving the entire NH 2 ( N424) or the entire COOH ( C92) terminus and examined the effects of these truncations on targeting, cell membrane stability, and NBC1 activity. N424 and C92 targeted to the plasma membrane of HEK293 cells or to the basolateral membrane of opossum kidney (OK) cells at 24 h but did not display NBC1 activity. Unlike the NBC1 wild-type and the N424, C92 expression was significantly decreased in the basolateral membrane at 48 h and yet the total C92 expression in the cell was constant. We found that decreased C92 expression in the basolateral membrane was due to increased endocytosis and mistargeting to the apical membrane. Increased endocytosis was prevented when both N424 and C92 were cotransfected together and more stable expression of C92 was observed. Immunoprecipitation studies using NBC1 antibody specific for the COOH epitope were able to detect the COOH truncated NBC1 when probed with NH 2 epitope-specific antibody or vice versa. Similar findings were observed with Ni-NTA pull-down assay. Cotransfection of both mutants partially restored NBC1 activity. In summary, NBC1 targets to the basolateral membrane of OK cells by a default mechanism and the COOH terminus plays a role on NBC1 stability in the basolateral membrane.
【关键词】 endocytosis basolateral membrane apical membrane proteinprotein interaction
THE SODIUM BICARBONATE COTRANSPORTER (NBC1) mediates bicarbonate reabsorption in the renal proximal tubule and is important for pH regulation in different cell types ( 2, 3, 21 ). Acute and chronic regulations of NBC1 have been studied extensively ( 7, 18, 22, 23 ). However, the molecular determinants necessary for understanding NBC1 activity in normal and disease states are not known.
NBC1 is localized in the basolateral membrane of the renal proximal tubule cells. NBC1 has 10 transmembrane-spanning domains with a long NH 2 and a short COOH terminus located intracellularly ( 31 ). Previous studies of other members of the bicarbonate superfamily, anion exchangers (AEs), have provided insights into the role of the NH 2 and the COOH termini on structure and function of the exchangers ( 14, 28, 33, 37, 39 ). In these studies, the NH 2 and COOH termini play an important role in regulation, localization, dimerization, and function of AEs. The NH 2 terminus of NBC1 contains conserved amino acids found in anion exchanger 2 (AE2) necessary for pH sensing ( 28 ) and contains a stretch of conserved amino acids found in AE1 necessary for activity ( 14 ). The NH 2 terminus of AE1 is also necessary for interaction with other cytoplasmic loops of AE1 in erythrocytes ( 14 ). Furthermore, crystallographic structure analysis of the NH 2 terminus of AE1 showed that its NH 2 terminus is important for cytoskeletal protein binding and dimerization ( 37 ) and is necessary for membrane insertion and transport activity ( 14 ).
The COOH terminus of NBC1 is very hydrophilic and is believed to be involved in protein-protein interaction and targeting to the plasma membrane ( 20 ). It contains 15 consecutive positively charged residues, 12 of which are lysine. It also contains an acidic motif, LDSDNDD, and a PKA phosphorylation site ( 8 ).
Toye et al. ( 33 ) reported that the COOH terminus of AE1 is involved in trafficking to the plasma membrane and further showed that the last 11 amino acids at the COOH terminus of the AE1 containing a tyrosine residue are responsible for basolateral targeting but are not sufficient for basolateral localization ( 32 ). Recently, Li et al. ( 17 ) showed that truncation of the last 23 amino acids of the COOH terminal caused mistargeting of NBC1 to the apical membrane at 48 h after transfection in Madin-Darby canine kidney (MDCK) cells. By contrast, Horita et al. ( 10 ) reported that deletion of 65 bp of SLC4A gene did not affect NBC1 function when expressed in ECV304 cells and in oocytes. This mutant, however, is retained in the intracellular compartment when expressed in MDCK cells. The reason for the apparent differences between the data of Li et al. ( 17 ) and Horita et al. ( 10 ) remained to be clarified.
The role of the NH 2 terminus in targeting, cell membrane expression, and NBC1 activity is not known. Although there is no clear basolateral targeting motif, the NBC1 NH 2 terminus contains a stretch of acidic amino acids that could potentially contribute to basolateral targeting and activity. In addition, several mutations of NBC1 in the NH 2 terminus resulted in impaired bicarbonate reabsorption leading to proximal renal tubular acidosis ( 12, 13 ), suggesting an important role of the NH 2 terminus in NBC1 function.
In the present study, we investigated whether the NH 2 and/or the COOH terminus plays a role in NBC1 activity, targeting, stability, and expression in the plasma membrane of a nonpolarized cell and in the basolateral membrane of polarized cells.
MATERIALS AND METHODS
Materials. The V5 antibody, PCR, cloning, and cell culture reagents including serum, lipofectamine 2000, and Ni-NTA agarose were obtained from Invitrogen (Grand Island, NY). The COOH- terminal-specific polyclonal NBC1 antibody was generated by ADI (San Antonio, TX). The NH 2 -terminal-specific polyclonal antibody was obtained from Chemicon (Temecula, CA). The pH-sensitive BCECF-AM was purchased from Molecular Probes (Eugene, OR). Protease inhibitor cocktail, phosphatase cocktail inhibitors I and II, amiloride, nigericin, and other perfusing reagents were obtained from Sigma (St. Louis, MO). Sulfo-NHS-LC-biotin, sulfo-NHS-SS-biotin, and streptavidin were purchased from Pierce (Rockford, IL). Immunoblots were analyzed with a chemiluminescence detection system (Phototype) from New England Biolabs (Beverly, MA). All other Western blotting solutions were obtained from Bio-Rad (Hercules, CA). NBC1 plasmid (pCI-NeoNBC1) was a gift from Dr. M. Soleimani (University of Cincinnati).
Construction of NBC1 truncation mutant. The NH 2 and/or the COOH termini of NBC1 ( Fig. 1 ) were truncated from full-length human NBC1 cloned in pCI-NeoNBC1 using PCR. The region of interest was amplified using primers upstream or downstream of the region to be deleted and cloned directionally in pcDNA3.1D/V5-His-TOPO vector (Invitrogen). Forward primers were designed to include Kozak sequence CACCATG at the 5'-end to ensure directionality. For V5-His-tagged constructs, the reverse primers were designed so the ORF is in frame with the COOH-terminal tag and did not include a stop codon. For untagged constructs, stop codon was included in the primers to prevent translation of the COOH-terminal V5/His tag found in the vector. Primers used for truncation of NBC1 are summarized in Table 1. A high-fidelity Pfu polymerase was used for DNA amplification to minimize errors, and constructs were sequenced using ABI automated sequencing (ABI PRISM 3100 Genetic Analyzer, Applied Biosystems, Foster City, CA). All plasmids were purified using Endofree plasmid purification kit (Qiagen).
Fig. 1. Schematic diagram of NBC1 truncations. N424 is a mutant with the entire NH 2 terminus truncated. N424 truncation was constructed by amplifying the NBC1 segment starting at the exact junction of the TM1. C92 is a mutant with the entire COOH terminus truncated, constructed by amplifying the NBC1 segment ending at the last NH 2 acid of the last transmembrane domain. TM is a mutant with both the NH 2 and the COOH termini truncated. Amplified DNA was cloned onto pcDNA3.1D/V5-His-TOPO vector. Mutants were tagged with V5-His at the COOH terminal and were used in most experiments. Untagged mutants were also generated and used as necessary. The V5-His tag did not affect NBC1 activity or localization to the plasma or basolateral membrane. WT, wild-type.
Table 1. Primers used for NBC1 truncation
Cell culture and transfection. Mycoplasma-free human embryonic kidney cells (HEK293) and opossum kidney (OK) cells were obtained from the American Type Culture Collection (Rockville, MD). Cells were maintained in 5% fetal bovine serum-supplemented DMEM/F12 (1:1) modified with penicillin, streptomycin, and 20 mM HEPES in a 5% CO 2 incubator at 37°C. For transfection, cells were grown to 95% confluency and transfected with equivalent amounts of NBC1 constructs using Lipofectamine 2000.
Intracellular pH i measurement and NBC1 activity. HEK293 cells were grown to 95% confluency and transiently transfected with molar equivalent amounts ( 1-1.1 µg) of WT-V5, N424-V5, C92-V5, TM-V5, or N424-V5/ C92-V5. After 24 h of transfection, cells were loaded with 6 µM BCECF-AM. pH-dependent changes in fluorescence were measured using a Deltascan dual-excitation fluorometer (Photon Technology International, Ontario, Canada) at 37°C, as previously described ( 23 ). This technique has been extensively used in our laboratory and assessed NBC activity accurately ( 1, 5, 35 ). In brief, cells were perfused at 37°C with Cl-free physiological solution (in mM): 25 NaHCO 3 -, 110 sodium gluconate, 5 potassium gluconate, 2 CaSO 4, 0.5 MgSO 4, 1 KH 2 PO 4, 10 glucose, and 9 HEPES, pH 7.40, containing 1 mM amiloride. After a stable baseline signal was established, Na + was removed by equimolar substitution with choline at pH 7.4. Na + removal resulted in an immediate decrease in pH i that fully recovered when Na + was readded. NBC1 activity was taken as an initial rate of pH recovery after the addition of Na + and was calculated from the slope of the line-drawn tangent to the initial deflection for a period of 1 min. All measurements were performed by dual-wavelength monitoring and ratiometric analysis at the pH-sensitive (504 nm) and pH-insensitive (440 nm) excitation wavelength depending on the BCECF-AM batch. During the measurement of NBC1 activity, cells were perfused at a constant extracellular pH of 7.4 and gassed with 5% CO 2. Although the buffering capacity was not measured in the present experiments, we previously showed that buffering capacity did not change during the course of the experiments.
Assessment of WT and truncated NBC1 localization to the plasma and to the basolateral membrane using cell-surface biotinylation: plasma membrane. HEK293 cells were grown on fibronectin-coated plates until 95% confluency and transfected with molar equivalent amounts of WT-V5, N424-V5, C92-V5, and TM-V5 using Lipofectamine 2000. Twenty-four hours after transfection, cells were labeled with sulfo-NHS-LC-biotin, as described by Sargiacomo et al. ( 25 ), with some modifications. This technique has been extensively used for measuring basolateral or apical membrane expression ( 6, 9, 36 ). In brief, cells were washed with PBS (pH 7.4) and labeled with 0.5 mg/ml sulfo-NHS-LC-biotin in PBS containing Ca 2+ /Mg 2+ for 25 min at 4°C. After 25 min of biotinylation, cells were washed with PBS and excess biotin was quenched with 100 mM glycine in PBS-Ca 2+ /Mg 2+. After labeling, cells were lysed using cell lysis buffer (1% Triton X-100, 150 mM NaCl, 5.0 mM EDTA, and 50 mM Tris, pH 7.5), supplemented with protease and phosphatase inhibitor cocktails. Biotinylated proteins were precipitated with immobilized streptavidin overnight. Biotin-labeled proteins were eluted by boiling the protein-biotin-streptavidin complex with 50 µl of elution buffer containing 80 mM DTT, 0.24 mM Tris·HCl, pH 8.9, 0.008% bromophenol blue, 5.6% SDS, and 16% glycerol. The proteins were subjected to SDS-PAGE, transferred onto a nitrocellulose membrane, and blotted with anti-NBC1 or anti-V5 antibody.
Basolateral membrane labeling. OK cells were grown on 0.45-µm polycarbonate membranes (Transwell) until 95% confluency and transfected with molar equivalent amounts of WT-V5, N424-V5, C92-V5, and TM-V5 using Lipofectamine 2000. The basolateral membrane was selectively labeled with 0.5 mg/ml sulfo-NHS-LC-biotin in PBS-Ca 2+ /Mg 2+ while the apical membrane was quenched with 100 mM glycine in PBS-Ca 2+ /Mg 2+. After labeling, cells were thoroughly washed with PBS, lysed, and biotinylated proteins were analyzed as described in plasma membrane labeling.
NBC1 localization time course. We employed a transient transfection technique to determine the time needed for newly synthesized NBC1 protein to localize in the basolateral membrane and to determine how long NBC1 stays in the membrane. OK cells were grown to 95% confluency and transfected with molar equivalent amounts of WT-V5, N424-V5, and C92-V5. Expression of the NBC1 truncated mutants in the basolateral membrane was assessed by labeling the basolateral membrane with biotin after 0, 6, 12, 24, and 48 h transfection, and biotin-labeled proteins were analyzed as described above.
Measurement of NBC1 endocystosis. The amount of endocytosed NBC1 was measured using 2-mercaptoethanesulfonic acid sodium salt (mesna) as previously described ( 6, 36 ). Mesna is a cell-impermeant agent capable of disrupting biotin-protein binding. Biotinylated proteins that remained in the basolateral membrane were exposed to mesna and are susceptible to cleavage, whereas biotinylated proteins that were endocytosed are protected. Resistance to mesna cleavage is used as a measure of endocytosis. OK cells grown on Transwells were transfected with molar equivalent amounts of WT-V5, N424-V5, and C92-V5. Twenty-four hours after transfection, the basolateral compartment was incubated with water-soluble biotin, 1.5 mg/ml sulfo-NHS-SS-biotin. After biotin labeling, unbound biotin was quenched with 100 mM glycine and the cells were washed with PBS-Ca 2+ /Mg 2+. Fresh complete media was added to the biotin-labeled cells in the presence or absence of 10 µM lactacystin (proteosomes inhibitor) and 20 mM chloroquine (lysosome inhibitor), and cells were reincubated in 5% CO 2 at 37°C for 24 h. After 24 h, cells were washed with cold PBS and were exposed to 1 ml of freshly prepared 10 mM mesna for 30 min. After 30 min, an additional 0.25 ml of 50 mM mesna was added. Another 0.25 ml of 50 mM mesna was added after 60 min. After 90 min of mesna incubation, mesna was oxidized by adding 1 ml of 500 mM iodoacetic acid for 10 min. Cells were lysed and biotinylated proteins were analyzed as described.
Immunoprecipitation studies of NH 2 and COOH truncated NBC1 mutants. HEK293 cells were grown on fibronectin-coated plates until 95% confluency and cotransfected with equal amounts of N424-V5 and C92-V5. Twenty-four hours after transfection, cells were lysed using modified RIPA buffer (50 mM Tris·HCl, pH 7.4, 1% NP 40, 150 mM NaCl, and 1 mM EDTA) supplemented with protease and phosphatase inhibitor cocktails. Cell lysates were precleared using protein A/G agarose for 10 min. Precleared lysates (500 µg) were incubated with anti-NBC1 antibody directed against the 53 amino acids within the cytoplasmic NH 2 terminus for 6 h. The resulting immunocomplex was precipitated with protein A/G agarose for 2-4 h. Proteins were eluted from the immunocomplex by boiling for 10 min in 2 x SDS sample buffer. Eluted proteins were resolved by denaturing PAGE, transferred onto nitrocellulose membrane, and probed with anti-NBC1 antibody directed against the 20 amino acids at the COOH terminus and vice versa.
Ni-NTA-agarose pull-down assay. Ni 2+ has the ability to bind to histidine. Polyhistidine-tagged protein binds to immobilized Ni 2+ -NTA complex with a K d of 10 -13 and is an important tool for protein purification. We employed this technique to determine whether His-tagged N424 pulls down the untagged C92. HEK 293 cells were grown on 100-mm plate and transfected with N424-V5-His. Twenty-four hours after transfection, cells were lysed using a nondenaturing lysis buffer (50 mM HEPES, pH 7.0, 0.5 M NaCl, 0.25% NP-40, 5 mM EDTA, and 10 mM imidazole) supplemented with protease and phosphatase inhibitor cocktails. One milligram of lysate was added to 50 µl of Ni-NTA-agarose and incubated for 1 h at 4°C. The Ni-NTA-agarose-His protein complex was incubated with wash buffer containing 20 mM imidazole for 5 min, three times. The washed Ni-NTA-agarose-His protein complex was incubated with 1 mg of untagged C92-transfected HEK293 cell lysate. Six hours after incubation, the Ni-NTA-agarose-His protein complex was washed thoroughly and bound protein/s were eluted using elution buffer containing 250 mM imidazole, pH 6.0. Eluted proteins were subjected to SDS-PAGE and Western blot analysis.
Statistical analysis. Densitometric analysis of Western blots was carried out using National Institutes of Health Image Analysis software. All data are expressed as means ± SE and statistical comparison using Student's t -test of paired or unpaired experiments, whenever appropriate. P values of <0.05 were considered to indicate significant difference.
RESULTS
NH 2 and COOH termini are not necessary for NBC1 targeting to the plasma membrane or to the basolateral membrane. NBC1 has a long NH 2 terminus (424aa) and a short COOH terminus (92aa) located intracellularly. To evaluate the role of the NH 2 and/or COOH terminus in targeting NBC1 to the plasma membrane, we truncated the whole NH 2 and/or COOH terminus ( Fig. 1 ). NBC1 truncation mutants were transfected onto HEK293 cells. HEK293 cells have no endogenous NBC1 expression and activity ( 3 ). Plasma membrane localization of NBC1-WT and NBC1 truncation mutants was assessed after 24 h using surface biotinylation. Figure 2 A, top, shows that NBC1-WT was transiently expressed in HEK293 cells and targeted to the plasma membrane as the endogenous NBC1 in OK cells and human proximal tubule kidney (HK2) cells. To validate the surface labeling technique, a Na + -K + -ATPase antibody was used to reprobe the membrane and Na + -K + -ATPase protein was detected ( Fig. 2 A, bottom ). Truncation of either the NH 2 or the COOH terminus did not prevent targeting and localization to the plasma membrane ( Fig. 2 B ). To examine whether the NH 2 or the COOH terminus influences targeting to the basolateral membrane, we used polarized OK cells grown on semipermeable membrane supports (Transwell) and transfected them with NBC1 truncation mutants. To ascertain localization in the basolateral membrane, 24 h after transfection, the basolateral membrane was labeled with biotin, lysed, precipitated with streptavidin, subjected to SDS-PAGE, and probed with anti-V5 antibody. Truncation of the NH 2 or the COOH terminus did not disrupt NBC1 localization to the basolateral membrane, but there was a smaller amount of NBC1 in the basolateral membrane when the COOH terminus was truncated. Truncation of both the NH 2 and the COOH termini caused a further decrease in NBC1 protein in the basolateral membrane ( Fig. 2 C ). Using the same nitrocellulose membrane, Na + -K + -ATPase, a marker of the basolateral membrane, was detected while sodium hydrogen exchanger 3 (NHE3), a protein that targets to the apical membrane, was not.
Fig. 2. NBC1 truncation mutants target to the plasma or to the basolateral membrane. Plasma membrane: HEK293 cells were grown in fibronectin-coated 6-well plates and transfected with molar equivalent amount of NBC1 constructs. Twenty-four hours after transfection, the plasma membrane expression of wild-type NBC1 and truncated NBC1 was assessed using cell-surface biotinylation. A, top : representative blot of 3 experiments showing that NBC1 antibody specific to the last 18 amino acids at the COOH terminus was able to detect the transfected wild-type NBC1 (WT) in the plasma membrane of HEK293 cells. Expression of endogenous NBC1 in HK2 and opossum kidney (OK) cells was used as positive control. The nitrocellulose membranes were stripped and reblotted with MAPK1/2 antibody as intracellular marker ( middle ), stripped for the second time, and reblotted with Na + -K + -ATPase antibody as plasma membrane marker ( bottom ). B, top : V5 antibody was able to detect the NH 2 truncated ( N424-V5), the COOH truncated ( C92-V5), and both the NH 2 and COOH (TM-V5) truncated NBC1 mutants in the plasma membrane. The anti-V5 antibody was not able to detect any protein in untransfected HEK293 cells lysate. Basolateral membrane : T o assess the basolateral membrane expression of the truncated NBC1 proteins, OK cells were polarized in semipermeable support, Transwell and transfected with molar equivalent amount of NBC1 constructs. Twenty-four hours after transfection, NBC1 expression in the basolateral membrane was assessed using surface biotinylation. C : truncation of the NH 2 and/or the COOH termini did not prevent localization of NBC1 to basolateral membrane. NHE3 antibody was used to reprobe the membrane to detect apical protein contamination ( middle ) and Na + -K + -ATPase antibody was used as control for equal protein loading and basolateral localization.
Truncation of the NH 2 and/or the COOH terminus disrupts NBC1 activity. Truncated NBC1 proteins were detected in the plasma membrane at 24 h. To determine whether any of these truncation mutants exhibits NBC activity, HEK293 cells were transfected with WT-V5, C92-V5, N424-V5, and TM-V5 and NBC1 activity was measured. Figure 3 shows that untransfected HEK293 cells showed no virtual pH recovery ( pH/min = 0.08 ± 0.03) while cells transfected with WT-V5 showed intracellular pH (pH i ) recovery ( pH/min = 0.51 ± 0.17) attributed to NBC1 activity. Expression of NBC1-truncated mutants in HEK293 cells did not exhibit pH recovery, indicating that the NH 2 and the COOH termini are necessary for NBC1 activity C92-V5 ( pH/min = 0.09 ± 0.06), N424-V5 ( pH/min = 0.08 ± 0.03), and TM-V5 ( pH/min = 0.07 ± 0.02). Figure 3 A is a representative tracing of (pH i ) recovery and Fig. 3 B shows the summary of eight experiments.
Fig. 3. NBC1 truncation mutants disrupt NBC1 activity. Quiescent HEK293 cells transiently expressing WT-V5, C92-V5, N424-V5, and TM-V5 were loaded with BCECF-AM. NBC1 activity was measured as the initial rate of pH recovery after sodium readdition using fluorometric analysis. A : superimposed representative tracing of intracellular pH recovery showing that HEK293 cells transiently expressing NBC1-WT recovered on sodium addition. The same rate of pH recovery was not observed in untransfected cells and cells overexpressing NBC1 truncation mutants. B : graphical representation of NBC1 activity expressed as pH/min. Data are shown as means ± SE of 8 experiments (*untransfected vs. WT-V5, P < 0.0008).
NH 2 -terminal truncation enhances NBC1 expression in the basolateral membrane while the COOH-terminal truncation decreases NBC1 expression in the basolateral membrane. Although all truncation mutants were detected in the plasma membrane or in the basolateral membrane at 24 h, the amount of truncated NBC1 proteins varied ( Fig. 2, B and C ). To determine whether this variation was due to NBC1 turnover in the plasma membrane, OK cells were transfected with molar equivalent amounts of WT-V5, N424-V5, or C92-V5 and the basolateral membrane was labeled with biotin at 0, 6, 12, 24, and 48 h after transfection. The WT-V5 was first expressed in the basolateral membrane 12 h after transfection, reached a steady-state level at 24 h, and remained stable at 48 h. Truncation of the NH 2 terminus ( N424-V5) was expressed in the basolateral membrane at 6 h after transfection and reached a steady-state level at 24 h. Truncation of the COOH terminus ( C92-V5) was expressed in the basolateral membrane at 12 h and reached steady state at 24 h. Unlike WT-V5 and N424-V5 expression in the basolateral membrane, C92-V5 expression decreased at 48 h, but total protein expression was unchanged ( Fig. 4 ).
Fig. 4. Truncation of the COOH terminus decreases NBC1 expression in the basolateral membrane of OK cells. OK cells were grown to 95% confluency on semipermeable supports (Transwell) and transfected with molar equivalent amount of WT-V5, N424-V5, and C92-V5 constructs. A : protein expression in the basolateral membrane was measured using surface biotinylation technique, 0, 6, 12, 24, and 48 h after transfection. Truncation of the COOH terminus ( C92-V5) resulted in decreased basolateral membrane expression at 48 h. Decreased basolateral membrane expression ( B ) was not observed with WT-V5 and N424-V5 ( left ). However, the C92-V5 expression in the total lysate (T) did not change ( right ). B : densitometric analysis showing that the expression of C92-V5 in the basolateral membrane is significantly decreased. Data shown as means ± SE of 4 paired experiments [* C92-V5 (B) vs. C92-V5 (T), P < 0.0002, # C92-V5 (B) vs. WT-V5 (B), P < 0.0003, C92-V5 (B) vs. N424-V5 (B), P < 0.0003].
Truncation of the COOH terminus increases endocytosis and causes mistargeting to the apical membrane in polarized OK cells. To define the mechanism of decreased C92-V5 expression in the basolateral membrane at 48 h, we measured endocytosis in polarized OK cells transfected with NBC1 mutants. Endocytosis was measured using a mesna protection assay in the presence or absence of chloroquine and lactacystin to prevent lysosomal and proteosomal degradation, respectively. Figure 5 shows that 24 h after labeling, a small amount of WT-V5 was endocytosed. The NBC1 wild-type that was endocytosed was completely degraded. The same result was observed with N424-V5. Truncation of the COOH terminus ( C92-V5) caused increased NBC1 endocytosis, but unlike the wild-type and the NH 2 -terminal truncated NBC1, some C92-V5 was not degraded. To study whether the undegraded, endocytosed C92-V5 was mistargeted to the apical membrane, we labeled the apical membrane 0, 6, 12, 24, 48 h after transfection. We also measured the amount of NBC1 left in the intracellular and the basolateral fraction (supernate). Figure 6 shows that while WT-V5 and N424-V5 were found in the basolateral and intracellular fraction, a significant amount of C92-V5 was found in the apical membrane at 24 h and the amount of mistargeted NBC1 increased at 48 h.
Fig. 5. Truncation of the COOH terminus increases endocytosis of NBC1 in polarized OK cells. OK cells were grown on semipermeable supports (Transwell) and transfected with equivalent amount of NBC1 constructs. A : twenty-four hours after transfection, the basolateral membrane was labeled with sulfo-NHS-SS-biotin, washed with PBS-Ca 2+ /Mg 2+, and reincubated with fresh OK medium in 5% CO 2 incubator at 37°C in the presence or absence of chloroquine and lactacystin (degradation pathway inhibitors) to allow endocytosis and to prevent degradation. Twenty-four hours after reincubation, the basolateral membrane was incubated with mesna to cleave biotin-protein binding. The amount of biotin-labeled NBC1 detected after mesna incubation was taken as a measure of endocytosis. Truncation of the COOH terminus caused increased endocytosis compared with WT-V5 and N424-V5. The endocytosed WT-V5 and N424-V5 were all degraded in the absence of cloroquine and lactacystin (-Clor/Lac), but a significant amount of C92-V5 was not. Blots shown are representative of 3 experiments. B : quantitative analysis showing that truncation of the COOH terminus increased endocytosis and significant amount of the endocytosed COOH truncated NBC1 was not degraded. Data are shown as means ± SE of 3 paired experiments (*endocytosed WT vs. endocytosed C92-V5, P < 0.006, #endocytosed N424-V5 vs. endocytosed C92-V5, P < 0.01, undegraded WT vs. undegraded C92-V5, P < 0.005, ¶undegraded N424-V5 vs. undegraded C92-V5, P < 0.004).
Fig. 6. Truncation of the COOH terminus causes mistargeting of NBC1 to the apical membrane of polarized OK cells. OK cells were grown on semipermeable supports (Transwell) and transfected with equivalent amounts of NBC1 constructs. Left : results of time course experiments showing that C92-V5 was mistargeted to the apical membrane after 24 h. Right : amount of NBC1 proteins in the supernate. The supernate contains all unlabeled proteins (basolateral +intracellular).
Cotransfection of the NH 2 and COOH truncated mutants prevents C92-V5 endocytosis and stabilizes NBC1 expression in the basolateral membrane. Since both the NH 2 and the COOH truncated NBC1 targeted to the basolateral membrane, we assessed whether expression of both truncation mutants prevents endocytosis of C92-V5. OK cells were cotransfected with C92-V5 and N424-V5 and membrane expression was measured after 0, 6, 12, 24, and 48 h. Figure 7 A, top, shows that when C92-V5 and N424-V5 were cotransfected, C92-V5 remained in the basolateral membrane at 48 h, whereas transfection of C92-V5 alone showed decreased expression. Figure 7 B is a densitometric analysis of C92-V5 expression showing increased C92-V5 protein stability in the basolateral membrane when coexpressed with N424-V5 as compared when C92-V5 was expressed alone in OK cells.
Fig. 7. N424-V5 stabilizes C92-V5 in the basolateral membrane. Equivalent amounts of N424-V5 and C92-V5 were cotransfected in polarized OK cells. The expression of truncated NBC1 proteins in the basolateral membrane was measured using surface biotinylation. A, top : cotransfection of N424-V5 and C92-V5 prevented the observed decrease of C92-V5 protein expression in the basolateral membrane at 48 h. Middle and bottom : timed expression of C92-V5 and N424-V5 in the basolateral membrane when expressed separately as already shown in Fig. 4 A. (These blots are used in this figure for the purpose of comparison.) Blots depicted are representative of 3 experiments. B : quantitative analysis of the blots showing stabilization of C92-V5 in the basolateral membrane when contransfected with N424-V5. Data shown as means ± SE of 3 paired experiments ( C92-V5 vs. C92-V5 + N424-V5, *24 h, P < 0.04, **48 h P < 0.007).
Coexpression of N424-V5 and C92-V5 restores NBC1 activity in HEK293 cells. Cotransfection of N424-V5 and C92-V5 stabilized protein expression in the basolateral membrane. To assess whether the NH 2 and the COOH termini can be pulled down by the appropriate antibody, N424-V5 and C92-V5 were cotransfected in HEK293 cells and the cell lysates were exposed to the antibody specific to either the NH 2 or the COOH epitope. Figure 8 A is a schematic diagram showing the epitopes of the antibodies used. Figure 8 B shows that an NBC1 antibody specific to the COOH-terminal epitope was able to detect the NH 2 -truncated NBC1 protein precipitated with NH 2 -specific NBC1 and vice versa. This shows that the NH 2 -truncated NBC1 was pulled down by the COOH truncated NBC1.
Fig. 8. NH 2 - and the COOH-terminal truncated mutants pull down each other. A : schematic diagram of the NBC1 epitopes specific for the antibodies used. B : N424-V5 and C92-V5 were cotransfected in HEK293 cells. Twenty-four hours after transfection, cells were lysed and NBC1 protein was immunoprecipitated with an antibody specific for the NH 2 terminus and probed with an antibody specific for the COOH terminus and vise versa. Left : NBC1 antibody specific for the COOH-terminal epitope -NBC1(C-term) was able to detect the NH 2 -truncated NBC1 protein which was pulled down using the NH 2 terminus-specific NBC1 antibody -NBC1 (N-term). Right : -NBC1 (N-term) was able to detect the COOH-truncated NBC1 protein pulled down by the COOH terminus-specific -NBC1 (C-term). These show that the NH 2 -truncated NBC1 coprecipitated with the COOH truncated NBC1 or vice versa. C : HEK293 cells were transfected with N424-V5-His. Twenty-four hours after transfection, cells were lysed and N424-V5-His was purified using Ni-NTA-agarose nondenaturing purification. Purified N424-V5-His-Ni-NTA agarose was incubated with untagged C92-transfected HEK293 cell lysates. Six hours after incubation, the N424-V5-His-Ni-agarose complex was washed thoroughly and proteins were eluted using a high concentration of imidazole. N424-V5-His-Ni-NTA agarose was shown to pull down the untagged C92, showing that the N424-V5-His-Ni-agarose forms a complex with C92.
To validate the above results, we used a Ni-NTA agarose pull-down assay. In this method, a His-tagged N424-V5 was purified by Ni-NTA agarose. Figure 8 C shows that the untagged C92 was pulled down with His-tagged N424-V5 ( right lane). There was no nonspecific binding when the untagged C92 was incubated with Ni-NTA-agarose in the absence of His-tagged N424-V5 (result not shown).
The HEK293 cells expressing both the N424-V5 and C92-V5 truncation mutants showed significantly increased NBC1 activity ( pH/min = 0.34 ± 0.02) compared with untransfected HEK293 cells or cells expressing NBC1 truncation mutants separately, as shown in Fig. 9.
Fig. 9. Coexpression of N424-V5 and C92-V5 restores NBC1 activity. N424-V5 and C92-V5 were cotransfected in HEK293 cells. Twenty-four hours after transfection, cells were incubated with BCECF-AM and NBC1 activity was measured. A graphical representation of NBC1 activity showing that coexpression of N424-V5 and C92-V5 restored NBC1 activity. Results were shown as means ± SE of 8 experiments (*untransfected vs. NBC1-WT, P < 0.0008, ** vs. N424 + C92, P < 0.001).
DISCUSSION
NBC1 mediates the vectorial transport of Na + and HCO 3 - in the basolateral membrane of renal proximal tubule cells. NBC1 targeting, stability in the membrane, and degradation are essential for its function. The present study was aimed at understanding the mechanisms involved in NBC1 targeting and NBC1 stability in the plasma membrane of nonpolarized cells and in the basolateral membrane of polarized cells. Specifically, the role of the cytoplasmic COOH and the NH 2 termini of NBC1 on targeting, expression, and stability was studied in the plasma membrane of HEK293 cells and in the basolateral membrane of OK proximal tubule cells.
Our data show that truncation of the NH 2 terminus, N424, or truncation of the COOH terminus, C92, did not prevent targeting of NBC1 to the plasma membrane of HEK293 cell or to the basolateral membrane of OK cells at 24 h ( Fig. 2 ). Truncation of both termini, TM, resulted in a significant decrease in NBC1 expression in the plasma or basolateral membrane, suggesting that both termini are necessary for NBC1 targeting and localization to the basolateral membrane. Assay of NBC1 activities revealed that all the mutants were devoid of NBC1 activity ( Fig. 3 ). The functional effect of truncating the NH 2 and the COOH termini shows that the NH 2 and the COOH termini play a role on NBC1 activity. This is consistent with the previous finding of Kanki et al. ( 14 ) that a series of NH 2 -terminal truncations of a member of the bicarbonate superfamily band 3/AE resulted in loss of transport activity, with preservation of surface localization.
These results are apparently different from those of Li et al. ( 17 ) at 48 h and Horita et al. ( 10 ). Li et al. ( 17 ) and Horita et al. ( 10 ) studied the role of the last 23 amino acids of the COOH terminus on NBC1 function and targeting to the basolateral membrane. Both studies showed that mutant NBC1 with truncation of the last 23 amino acids remains functional when expressed in oocytes. Li et al. ( 17 ), however, showed mistargeting of this NBC1 mutant to the apical membrane with residual basolateral expression while Horita et al. ( 10 ) showed that the truncated NBC1 is retained in the intracellular compartments when expressed in MDCK cells.
By following the time course of total and basolateral NBC1 truncation mutant expression, we found that truncation of the NH 2 terminus increased its expression in the basolateral membrane after 6 h, reaching a maximum at 24 h, and remained stable throughout the 48 h. On the other hand, expression of the COOH truncated NBC1 mutant increased in the basolateral membrane up to 24 h but decreased at 48 h, while the total amount of COOH truncated NBC1 did not change ( Fig. 4 ). This decreased expression of the COOH truncated NBC1 in the basolateral membrane could be accounted for by either increased endocytosis and/or mistargeting to the apical membrane. Truncation of the COOH terminus of NBC1 caused increased endocytosis when measured in the presence or absence of lactacystin (proteosome inhibitor) and chloroquine (lysosome inhibitor). Unlike the wild-type and the NH 2 -truncated NBC1, some of the endocytosed COOH terminus was not degraded ( Fig. 5 ) but mistargeted to the apical membrane ( Fig. 6 ). This finding confirms the observation by Li et al. ( 17 ), in that there was a predominant apical membrane targeting of NBC1 at 48 h. The current results suggest that NBC1 initially targeted to the basolateral membrane, followed by increased endocytosis, which consequently increased the amount of truncated NBC1 in the intracellular pool followed by rerouting of the COOH-terminal trunctated NBC1 to the apical membrane. The present findings are also consistent with the increased amount of NBC1 found in the intracellular pool as described by Horita et al. ( 10 ). Other studies have also shown that same truncated protein can be expressed in either the apical or basolateral membrane of certain cells ( 29, 30 ). These differences in expression of different proteins may be related to the length of truncation (which may affect protein structure, folding, and targeting) ( 4, 15 ) and cell type used for expression.
Our study shows that the COOH-terminal truncated mutant remained in the basolateral membrane after 48 h ( Fig. 7, top ) when cotransfected with the NH 2 truncated NBC1, unlike what was observed when C92 and N424 were expressed separately ( Fig. 4 and 7, middle and bottom ). These results suggest increased stability of the COOH truncated NBC1 when both mutants were cotransfected. The role of the NH 2 and COOH termini interaction on protein folding, function, stability, and turnover has been the subject of a recent study. Krishna and Englander ( 15 ) showed that the NH 2 and the COOH termini of several transmembrane proteins interact and play a role in protein folding and stability. Immunoprecipitation experiments using an NBC1 antibody specific for the COOH terminus were able to detect the NH 2 -truncated NBC1 precipitated with NH 2 -specific NBC1 antibody and vice versa, showing that the NH 2 and the COOH termini coprecipitated ( Fig. 8 B ). The same result is shown using Ni-NTA-agarose pull down assay ( Fig. 8 C ).
Of great interest is the finding that cotransfection of the NH 2 - and COOH-terminal truncated mutants restored NBC1 activity ( Fig. 9 ). The mechanism for the restoration of function is not clarified by our study. Previous studies have shown that the NH 2 and the COOH termini of some proteins interact and that this interaction is neccesary for function as described for the inwardly rectifying potassium channels ( 34 ), the cyclic nucleotide gating channel ( 38 ), androgen receptor ( 11, 16 ), the mechanosensitive chloride channel ( 19 ), and the voltage-gated chloride channels ( 26 ). Our results could be explained either by interaction of the two truncated mutants or by the possibility that the NH 2 - and the COOH-terminal truncations share residues within the central hydrophobic core that are required for NBC1 dimerization ( 24, 27 ). Thus coexpression would bring together the NH 2 and the COOH termini in a significant proportion of dimers, resulting in correction of the mutant phenotypes. Our data do not allow us to distinguish between these two hypotheses.
In summary, the present study provides information on the regulation of NBC1 in the cell membrane. The NH 2 and COOH termini are necessary for NBC1 activity but are not required for targeting NBC1 protein to the basolateral membrane. The COOH terminus contains molecular determinants necessary to anchor NBC1 protein in the basolateral membrane. Truncation of the COOH terminus resulted in increased endocytosis followed by mistargeting to the apical membrane but is rescued in the presence of a mutant containing the NH 2 terminus.
GRANTS
This work was supported in part by a Merit Review Program Award from the US Department of Veterans Affairs to J. A. L. Arruda and A. A. Bernardo as well as by grants-in-aid from local affiliates of the National Kidney Foundation to A. A. Bernardo.
【参考文献】
Bernardo AA, Espiritu DJ, Ruiz OS, Robey RB, and Arruda JA. The role of phosphatidylinositol 3-Kinase (PI3K) in CO 2 stimulation of the Na + /HCO 3 - cotransporter (NBC). J Membr Biol 191: 141-148, 2003.
Boron WF and Boulpaep EL. Intracellular pH regulation in the renal proximal tubule of the salamander. Basolateral HCO 3 - transport. J Gen Physiol 81: 53-94, 1983.
Burnham CE, Amlal H, Wang ZH, Shull GE, and Soleimani M. Cloning and functional expression of a human kidney Na + /HCO 3 - cotransporter. J Biol Chem 272: 19222-19224, 1997.
Dobson CM. Protein folding and misfolding. Nature 426: 884-890, 2003.
Espiritu DJ, Bernardo AA, Robey RB, and Arruda JA. A central role of Pyk2-Src interaction in coupling diverse stimuli to increased epithelial NBC activity. Am J Physiol Renal Physiol 283: F663-F670, 2002.
Espiritu DJ, Yang V, Bernardo A, and Arruda JA. Regulation of renal Na + /HCO 3 - cotransporter stimulation by CO 2 : role of phosphorylation, exocytosis and protein synthesis. J Membr Biol 199: 39-49.
Giebel JP, Geibisch G, and Boron WF. Angiotensin II stimulates Na + /H + exchanger and Na/HCO 3 cotransport in rabbit proximal tubule. Proc Natl Acad Sci USA 87: 7917-7920, 1990.
Gross E, Hawkins K, Pushkin A, Sassani P, Dukkipati R, Abuladze N, Hopfer U, and Kurtz I. Phosphorylation of Ser(982) in the sodium bicarbonate cotransporter kNBC1 shifts the HCO 3 - :Na + stoichiometry from 3:1 to 2:1 in murine proximal tubule cells. J Physiol 537: 659-665, 2001.
Hanwell D, Ishikawa T, Saleki R, and Rotin D. Trafficking and cell surface stability of the epithelial Na+ Channel expressed in epithelial Madin-Darby Canine kidney cells. J Biol Chem 277: 9772-9779, 2002.
Horita S, Yamada H, Inatomi J, Sekine T, Igarashi T, Seki G, and Fugita T. Mechanism of NBC1 inactivation by mutations identified in patients with proximal renal tubular acidosis and ocular abnormalities (Abstract). J Am Soc Nephrol 16: 123A, 2005.
Hsu CL, Chen YL, Ting HJ, Yang Z, Zhang Y, Wang CT, Chang HC, Yeh S, Pimplikar SW, and Chang C. Androgen Receptor (AR) NH 2 - and COOH-terminal interactions result in the differential influences on the AR-mediated transactivation and cell growth. Mol Endocrinol 19: 350-361, 2005.
Igarashi T. Mutations in SLC4A4 cause permanent isolated proximal renal tubular acidosis with ocular abnormalities. Nature 23: 264-265, 1999.
Igarashi T, Inatomi J, Sekine T, Seki G, Yamada H, Horita S, and Fujita T. Mutational and functional analysis of the Na/HCO 3 cotransporter gene ( SLC4AC ) in patients with permanent isolated proximal renal tubular acidosis and ocular abnormalities. J Am Soc Nephrol 13: 302A, 2003.
Kanki T, Young MT, Sakaguchi M, Hamasaki N, and Tanner MJ. The N-terminal region of the transmembrane domain of human erythrocyte band 3. Residues critical for membrane insertion and transport activity. J Biol Chem 278: 5564-5573, 2003.
Krishna MM and Englander SW. The N-terminal to C-terminal motif in protein folding and function. Proc Natl Acad Sci USA 102: 1053-1058, 2005.
Langley E, Kemppainen JA, and Wilson EM. Intermolecular NH 2 -/Carboxyl-terminal interactions in androgen receptor dimerization revealed by mutations that cause androgen insensitivity. J Biol Chem 273: 92-101, 1998.
Li HC, Worrell RT, Matthews JB, Husseinzadeh H, Neumeier L, Petrovic S, Conforti L, and Soleimani M. Identification of a carboxyl-terminal motif essential for the targeting of Na + -HCO 3 - cotransporter NBC1 to the basolateral membrane. J Biol Chem 279: 43190-43197, 2004.
Mckinney TD and Myers P. Bicarbonate transport by proximal tubules: effect of parathyroid hormone and dibutyril cAMP. Am J Physiol Renal Fluid Electrolyte Physiol 238: F166-F174, 1980.
Park KH, Berrier C, Martinac B, and Ghazi A. Purification and functional reconstitution of N- and C-halves of the MscL channel. Biophys J 86: 2129-2136, 2004.
Pushkin A, Abuladze N, Gross E, Newman D, Tatishchev S, Lee I, Fedotoff O, Bondar G, Azimov R, Ngyuen M, and Kurtz I. Molecular mechanism of kNBC1-carbonic anhydrase II interaction in proximal tubule cells. J Physiol 559: 55-65, 2004.
Romero MF, Hediger MA, Boulpaep EL, and Boron WF. Expression cloning and characterization of a renal electrogenic Na + /HCO 3 - cotransporter. Nature 387: 409-413, 1997.
Ruiz OS, Arruda JA, and Talor Z. Na-HCO 3 cotransport and Na-H antiporter in chronic respiratory acidosis and alkalosis. Am J Physiol Renal Fluid Electrolyte Physiol 256: F414-F420, 1989.
Ruiz OS, Wang LJ, Pahlavan P, and Arruda JA. Regulation of renal Na-HCO 3 cotransporter. III. Presence and modulation by glucocorticoids in primary cultures of the proximal tubule. Kidney Int 47: 1669-1679, 1995.
Russ WP and Engelman DM. The GxxxG motif: a framework for transmembrane helix-helix association. J Mol Biol 296: 911-919, 2000.
Sargiacomo M, Lisanti M, Graeve L, Le Bivic A, and Rodriquez-Boulan E. Integral and protein composition of the apical and basolateral membrane domains in MDCK cells. J Membr Biol 107: 277-286, 1989.
Schmidt-Rose T and Jentsch TJ. Reconstitution of functional voltage-gated chloride channels from complementary fragments of CLC-1. J Biol Chem 272: 20515-20521, 1997.
Schoneberg T, Yun J, Wenkert D, and Wess J. Functional rescue of mutant V2 vasopressin receptors causing nephrogenic diabetes insipidus by a coexpressed receptor polypeptide. EMBO J 15: 1283-1291, 1996.
Stewart AK, Chernova MN, Shmukler BE, Wilhelm S, and Alper SL. Regulation of AE2-mediated Cl - transport by intracellular or by extracellular pH requires highly conserved amino acid residues of the AE2 NH 2 -terminal cytoplasmic domain. J Gen Physiol 120: 707-722, 2002.
Sun AQ, Swaby I, Xu S, and Suchy FJ. Cell-specific basolateral membrane sorting of the human liver Na + -dependent bile acid cotransporter. Am J Physiol Gastrointest Liver Physiol 280: G1305-G1313, 2001.
Swiatecka-Urban A, Duhaime M, Coutermarsh B, Karlson KH, Collawn J, Milewski M, Cutting GR, Guggino WB, Langford G, and Stanton BA. PDZ domain interaction controls the endocytic recycling of the cystic fibrosis transmembrane conductance regulator. J Biol Chem 277: 40099-40105, 2002.
Tatishchev S, Abuladze N, Pushkin A, Newman D, Liu WX, Weeks D, Sachs G, and Kurtz I. Identification of membrane topography of the electrogenic sodium bicarbonate cotransporter pNBC1 by in vitro transcription/translation. Biochemistry 42: 755-765, 2003.
Toye AM, Banting G, and Tanner MJ. Regions of human kidney anion exchanger 1 (kAE1) required for basolateral targeting of kAE1 in polarised kidney cells: mis-targeting explains dominant renal tubular acidosis (dRTA). J Cell Sci 117: 1399-1410, 2004.
Toye AM, Bruce LJ, Unwin RJ, Wrong O, and Tanner MJA. Band 3 Walton, a C-terminal deletion associated with distal renal tubular acidosis, is expressed in the red cell membrane but retained internally in kidney cells. Blood 99: 342-347, 2002.
Tucker SJ and Ashcroft FM. Mapping of the physical interaction between the intracellular domains of an inwardly rectifying potassium channel, Kir6.2. J Biol Chem 274: 33393-33397, 1999.
Weinman EJ, Evangelista CM, Steplock D, Liu MZ, Shenolikar S, and Bernardo A. Essential role for NHERF in cAMP-mediated inhibition of the Na + -HCO 3 - cotransporter in BSC-1 cells. J Biol Chem 276: 42339-42346, 2001.
Yang X, Amemiya M, Peng Y, Moe OW, Preisig PA, and Alpern RJ. Acid incubation causes exocytic insertion of NHE3 in OKP cells. Am J Physiol Cell Physiol 279: C410-C419, 2000.
Zhang D, Kiyatkin A, Bolin JT, and Low P. Crystallographic structure and functional interpretation of the cytoplasmic domain of erythrocyte membrane band 3. Blood 96: 2925-2933, 2000.
Zheng J, Varnum MD, and Zagotta WN. Disruption of an intersubunit interaction underlies Ca 2+ -calmodulun modulation of cyclic nucleotide-gated channels. J Neurosci 23: 8167-8175, 2003.
Zhu Q, Lee DWK, and Casey JR. Topology in C-terminal region of the human plasma membrane anion exchangers, AE1. J Biol Chem 278: 3112-3120, 2003.
作者单位：1 Section of Nephrology, Department of Medicine, and 2 Department of Physiology and Biophysics (M/C 901), University of Illinois at Chicago, Chicago; and 3 Jesse Brown Veterans Affairs Medical Center, Chicago, Illinois